Canine ischaemic dermatopathy: a retrospective study of 177 cases (2005–2016)
Abstract
enBackground
Ischaemic dermatopathy encompasses a poorly understood subset of canine diseases that share similar clinical and histological features. Very little information is currently available regarding population characteristics, progression and outcome.
Hypothesis/objectives
This study aimed to describe the clinical features and therapeutic outcomes of ischaemia dermatopathy, excluding familial dermatomyositis, using cases diagnosed by histopathological analysis.
Animals
One hundred and seventy-seven cases submitted for histopathological analysis between 2005 and 2016 met inclusion criteria, of which 93 had complete medical records available.
Methods and materials
Both records and pointed surveys were used to retrieve information. Scoring systems were created to subjectively evaluate clinical outcomes and likelihood of a vaccine association.
Results
Of 177 cases, toy and miniature poodles, Chihuahuas, Maltese, Yorkshire terriers and Jack Russell terriers were significantly over-represented (P < 0.001). Of the 93 cases for which historical data were obtained, median age at skin biopsy was five years (0.42–13 years) and median body weight was 7.3 kg (range 1.32–50.3 kg). The condition in 45 dogs (48.3%) was found likely to be associated with vaccination. Younger ages (P = 0.011) and higher body weights (P = 0.003) were positively correlated with greater likelihood of vaccination. Body weight <10 kg (P = 0.0045) and older ages (P = 0.0048) were significantly associated with worse outcomes.
Conclusions and clinical importance
This study provides support for breed predispositions and identifies potential prognostic factors. Importantly, over half of the cases were considered unlikely to be vaccine-associated, demonstrating the need to investigate other underlying causes of this condition.
Résumé
frContexte
La dermatopathie ischémique regroupe un ensemble de dermatoses peu connues du chien qui partagent des similitudes cliniques et histologiques. Très peu d'informations sont actuellement disponibles concernant les caractéristiques de la population, la progression et la guérison.
Hypothèses/Objectifs
Cette étude a pour but de décrire les données cliniques et thérapeutiques de la dermatopathie ischémique, à l'exclusion de la dermatomyosite familiale, à l'aide de cas diagnostiqués par analyse histopathologique.
Sujets
Cent soixante dix sept cas soumis pour analyse histopathologique entre 2005 et 2016 ont rempli les conditions d'inclusion, dont 93 avaient des données médicales complètes.
Matériel et méthode
Les données et les études référencées ont été utilisées pour extraire les informations. Des systèmes de notation ont été créés pour évaluer subjectivement les résultats cliniques et la probabilité d'une association au vaccin.
Résultats
Sur les 177 cas, les caniches nains et miniatures, chihuahuas, bichon maltais, Yorkshire terriers et Jack Russel terriers étaient significativement surreprésentés (P < 0.001). Sur les 93 cas dont les commémoratifs ont été obtenus, l’âge médian au moment des biopsies cutanées était de cinq ans (0.42–13 ans) et le poids moyen était de 7,3 kg (1.32–50.3 kg). Pour 45 chiens (48.3%), une association avec un vaccin était très probable. Un jeune âge (P = 0.011) et un poids élevé (P = 0.003) étaient corrélés positivement avec un plus grand risque associé à la vaccination. Un poids <10 kg (P = 0.0045) et un âge plus avancé (P = 0.0048) étaient significativement associés à des atteintes plus sévères.
Conclusions et importance clinique
Cette étude fournit un support pour les prédispositions raciales et identifie les facteurs pronostics potentiels. En outre, plus de la moitié des cas étaient considérés comme peu susceptibles d’être associés à la vaccination, démontrant le besoin d'explorer les autres causes possibles de cette dermatose.
Resumen
esIntroducción
la dermatopatía isquémica abarca un subconjunto poco conocido de enfermedades caninas que comparten características clínicas e histológicas similares. Actualmente hay muy poca información disponible sobre las características de la población, la progresión y el resultado.
Hipótesis/objetivos
este estudio tuvo como objetivo describir las características clínicas y los resultados terapéuticos de la dermatopatía isquémica, excluyendo la dermatomiositis familiar, utilizando casos diagnosticados mediante análisis histopatológico.
Animales
ciento setenta y siete casos presentados para el análisis histopatológico entre 2005 y 2016 cumplieron con los criterios de inclusión, de los cuales 93 tenían historiales médicos completos disponibles.
Métodos y materiales
tanto los historiales como las encuestas selectivas se utilizaron para recopilar información. Se crearon sistemas de puntuación para evaluar subjetivamente los resultados clínicos y la probabilidad de una asociación con vacunas.
Resultados
de 177 casos, Caniches Toy y Miniatura, Chihuahuas, Malteses, Yorkshire Terriers y Jack Russell Terrier estaban significativamente sobre representados (P <0,001). De los 93 casos en los cuales se obtuvieron datos clínicos, la edad media en el momento de la biopsia de piel fue de cinco años (0,42 a 13 años) y el peso corporal promedio fue de 7,3 kg (rango de 1,32 a 50,3 kg). Se encontró que la condición en 45 perros (48.3%) estuvo asociada con la vacunación. Edades más jóvenes (P = 0.011) y pesos corporales más altos (P = 0,003) se correlacionaron positivamente con una mayor probabilidad de vacunación. El peso corporal <10 kg (P = 0,0045) y las edades mayores (P = 0,0048) se asociaron significativamente con peor evolución.
Conclusiones e importancia clínica
este estudio contribuye a demostrar la predisposición racial e identifica posibles factores pronósticos. Es importante destacar que en más de la mitad de los casos se consideró poco probable que hubiese asociación a vacunas, lo que demuestra que es necesario investigar otras causas subyacentes de esta enfermedad.
Zusammenfassung
deHintergrund
Die ischämische Dermatopathie umfasst eine wenig verstandene Untergruppe von Erkrankungen des Hundes, die ähnliche klinische und histologische Merkmale teilen. Zurzeit gibt es sehr wenig Information über Populationscharakteristika, Progression und Endergebnis.
Hypothese/Ziele
Diese Studie hatte zum Ziel die klinischen Charakteristika und die therapeutischen Ergebnisse einer ischämischen Dermatopathie zu beschreiben, wobei die familiäre Dermatomyositis ausgenommen war. Es wurden Fälle verwendet, die mittels histopathologischer Analyse diagnostiziert worden waren.
Tiere
Einhundertsiebenundsiebzig Fälle, die zwischen 2005 und 2016 zur histopathologischen Analyse gelangten, erfüllten die Einschlusskriterien. Von 93 Fällen stand die gesamte medizinische Patientenkartei zur Verfügung.
Methoden und Materialien
Sowohl die Patientenkarteien wie auch punktierte Surveys wurden verwendet um an Informationen zu gelangen. Es wurden Bewertungssysteme kreiert, um die klinischen Ergebnisse subjektiv zu evaluieren und die Wahrscheinlichkeit eines Zusammenhangs mit einer Impfung festzustellen.
Ergebnisse
Von 177 Fällen waren Toy- und Zwergpudel, Chihuahuas, Malteser, Yorkshire Terrier und Jack Russell Terrier signifikant überrepräsentiert (P < 0,001). Von den 93 Fällen von denen anamnestische Daten vorlagen, lag das mediane Alter zum Zeitpunkt der Hautbiopsie bei fünf Jahren (0,42-13 Jahre) und das mediane Körpergewicht bei 7,3 kg (1,32-50,3 kg). Es wurde für wahrscheinlich erachtet, dass die Erkrankung bei 45 Hunden (48,3%) mit einer Impfung im Zusammenhang stand. Die Erkrankung war bei jüngeren Hunden (P = 0,011) und solchen mit einem höheren Körpergewicht (P = 0,003) positiv korreliert mit einer höheren Wahrscheinlichkeit, dass eine Impfung damit im Zusammenhang stand. Ein Körpergewicht < 10 kg (P = 0,0045) und ältere Hunde (P = 0,0048) zeigten einen signifikanten Zusammenhang mit einem schlechteren Ausgang.
Schlussfolgerungen und klinische Bedeutung
Diese Studie unterstützt Rassenprädispositionen und identifiziert mögliche prognostische Faktoren. Wichtig ist festzuhalten, das bei über die Hälfte der Fälle kein Zusammenhang mit einer Impfung gesehen wurde, was auf die Wichtigkeit hinweist, auch andere zugrundeliegende Ursachen dieser Erkrankung zu erforschen.
要約
ja背景
虚血性皮膚症は、犬疾患における類似した臨床的および組織学的特徴を有する、あまり理解されていないサブセットを含む。現在は集団特性、進行過程および治療成果に関する情報はほとんどない。
仮説/目的
本研究は、病理組織学的解析によって診断した症例を用いて、家族性皮膚筋炎を除く、虚血性皮膚疾患の臨床的特徴および治療成果を記述することを目的とした。
被験動物
2005年から2016年の間に病理組織学的解析のために提出された177例が選択基準を満たし、そのうち93例は完全な医療記録を入手できた。
材料と方法
情報の検索に医療記録およびポイントを絞った調査の両方を利用した。スコアリングシステムを、臨床成果およびワクチンとの関連性の可能性を評価するために主観的に作成した。
結果
177例のうちトイおよびミニチュア・プードル、チワワ、マルチーズ、ヨークシャー・テリアおよびジャック・ラッセル・テリアは有意に過剰発現していた(P <0.001)。過去の医療記録を得た93例のうち、皮膚生検を実施した年齢中央値は5歳(0.42〜13歳)、体重中央値は7.3 kg(範囲:1.32〜50.3 kg)であった。 45頭(48.3%)の犬の臨床症状がワクチン接種と関連している可能性が高いことがわかった。より若い年齢(P = 0.011)およびより重い体重(P = 0.003)が、ワクチン接種に対するより高い正の相関を示した。体重<10 kg(P = 0.0045)およびそれ以上の年齢(P = 0.0048)は、より悪い治療成果と有意に関連していた。
結論と臨床的重要性
本研究は品種の素因を裏付け、潜在的な予後因子を特定している。重要なことに、半数以上の症例がワクチン関連ではなさそうであると考えられ、本疾患症状に対する他の根本的な原因を調査する必要性を示唆している。
摘要
zh背景
犬缺血性皮肤病包含一些临床和组织学特征相似且机制不明的犬病组群。目前关于种群特性、进展和结果的信息非常少。
假设/目标
本研究旨在通过组织病理学诊断的病例,报告缺血性皮肤病的临床特征和治疗结果(不包括家族性皮肌炎)。
动物
2005年至2016年期间,177个做过组织病理学分析且符合纳入标准的病例,其中93个病例有完整病历。
方法和材料
检索信息进行有针对性的调查并记录,创建评分系统去主观评估临床结果与疫苗相关的可能性。
结果
在177例中,玩具贵宾犬、迷你贵宾犬、吉娃娃犬、马耳他犬、约克夏㹴犬和杰克罗素㹴犬的比例明显较多(P <0.001)。在93例历史数据中,皮肤活检的中位年龄为5岁(0.42-13岁),中位体重为7.3kg(范围1.32-50.3kg)。从45只犬身上发现(48.3%)病情可能与疫苗接种有关。年龄越小(P = 0.011)和体重越高(P = 0.003),以至于接种疫苗的可能性越大,呈正相关。 体重<10 kg(P = 0.0045)和年龄越大(P = 0.0048),以至于预后越差,呈显著相关性。
结论和临床意义
该研究为品种易感性提供支持,并识别潜在的预后因素。重要的是,超过一半的病例与疫苗无关,这表明有必要调查该病的其他潜在原因。
Resumo
ptContexto
O termo dermatopatia isquêmica compreende uma categoria mal compreendida de doenças caninas que compartilham características clínicas e histopatológicas similares. Há poucas informações disponíveis sobre as características populacionais, progressão e evolução.
Hipótese/objetivos
Este estudo teve como objetivo descrever os aspectos clínicos e o resultado terapêutico da dermatopatia isquêmica, excluindo a dermatomiosite familiar, utilizando casos diagnosticados por exame histopatológico.
Animais
Cento e setenta e sete casos submetidos à análise histopatológica entre 2005 e 2016 entraram nos critérios de inclusão, dos quais 93 apresentaram prontuários completos.
Métodos e materiais
Ambos históricos e questionários foram utilizados para pesquisar informações. Foram criados sistemas de escore para avaliar subjetivamente as evoluções clínicas e a probabilidade de associação com vacinação.
Resultados
Dos 177 casos, os poodles toys e miniatura, Chihuahuas, Malteses, Yorkshire terriers e Jack Russell terriers estavam significativamente super-representados P < 0.001). Dos 93 casos em que foi possível se obter dados de histórico, a idade mediana na data da biópsia cutânea foi de cinco anos (0,42–13 anos) e o peso mediano foi de 7,3kg (1,32 – 50,3kg). A doença em 45 cães (48,3%) provavelmente estava associada à vacinação. Juventude (P = 0.011) e peso corpóreo mais elevado (P = 0.003) foram positivamente correlacionados com uma maior probabilidade de vacinação. Peso corpóreo <10kg (P = 0.0045) e idade avançada (P = 0.0048) foram significativamente associados a pior evolução.
Conclusões e importância clínica
Este estudo corrobora com predisposições raciais e identifica potenciais fatores prognósticos. É relevante ressaltar que em metade dos casos, foi considerada improvável a associação com vacinação, demonstrando a necessidade de se investigar outras causas primárias desta enfermidade.
Introduction
Ischaemic dermatopathy is a diagnosis that encompasses different clinical syndromes with common features and a wide range of possible aetiologies. Ischaemic diseases previously have been grouped into five subtypes: canine familial dermatomyositis (FDM), juvenile onset dermatomyositis-like disease in atypical breeds, post-rabies vaccine panniculitis, generalized vaccine-associated ischaemic dermatopathy and generalized idiopathic ischaemic dermatopathy.1
Whether generalized or localized, cutaneous signs include alopecia, hyper- or hypopigmentation, scaling, crusts and ulcers. Lesions are found most commonly at vaccine sites and/or generalized areas including the pinnal margin, tip of the tail, head, face and pressure points. An acral distribution also is frequently observed and characterized by lesions on phalanges, paw pads and claws (onychodystrophy and onychomadesis).
Histological features of ischaemic dermatopathy include follicular atrophy, a cell-poor lymphocytic interface dermatitis, dermal oedema/mucin deposition, and eosinophilic tinctorial changes to the dermal collagen and vascular tunics imparting a “smudged” appearance to the collagen. These changes are thought to represent a chronic oxygen-/nutrient-deficient state of the affected tissue.1, 2
Despite being recognized in the mid-1980s, very few primary publications since then have focused on nonfamilial variants of ischaemic dermatopathy.2-6 Although rabies vaccine-associated cases are relatively well recognized in clinical practice, there is a paucity of information regarding the true patient population, risk factors and clinical outcomes. Additionally, even less is known about idiopathic ischaemic dermatopathies in which an obvious vaccine trigger is not identified. Furthermore, optimal medical management of vaccine-associated and idiopathic cases has not been examined closely. Pentoxifylline has been reported to be effective in cases of FDM and is used for other subtypes of ischaemic dermatopathy, despite few published supportive data.7 Three cases of vaccine-associated disease were reported to respond well to therapy with pentoxifylline combined with vitamin E (and in two of these cases, short courses of prednisone).2 A report of two cases described excellent responses to oclacitinib.6 However, anecdotal reports suggest that these cases can be difficult to manage, with some requiring combination immunosuppressive therapy, and others progressing despite treatment.5, 8
The objectives of this study were to develop more accurate descriptions of the population characteristics, clinical presentations, and outcome features of the various subtypes of ischaemic dermatopathy, excluding FDM. Additionally, an effort was made to correlate clinical findings and treatment data with outcome patterns in order to identify risk factors and prognostic indicators for disease severity and progression. Based on clinical impressions, the authors hypothesized that small breeds would comprise the majority of the patient population, specifically of those dogs with vaccine-associated disease. Additionally, the authors hypothesized that dogs with a body weight <10 kg, and those without a strong vaccine association, would have worse outcomes.
Methods and materials
A comprehensive electronic search of a histopathology database at the School of Veterinary Medicine of the University of Pennsilvania for reports generated between 2005 and 2016 containing the key words “ischaemic” or “ischaemic dermatopathy”, was performed. Cases were then selected if the histological description and/or written diagnosis was consistent with cutaneous ischaemic disease. Exclusion criteria included a diagnosis of FDM (excluding all currently reported breeds) and ischaemic changes secondary to a primary skin disease (e.g. cutaneous neoplasia, reactive histiocytosis). A total of 282 cases were initially identified. All slides from each case were then examined by two authors.
A diagnosis of ischaemic dermatopathy was conferred by the presence of adnexal atrophy (“faded follicles”) with one or more of the following histological characteristics: (1) cell-poor interface pattern, (2) collagen smudging, (3) vasculitis or vasculopathy (loss of vascular distinction or endothelial cells) and/or (4) lymphoplasmacytic dermatitis or panniculitis. (Figure 1).
Photomicrographs showing histopathological features of ischaemic dermatopathy in skin from Case 101.
(a) Mild dermal inflammation and oedema with marked follicular atrophy (faded follicles) with a prominent glassy membrane (arrows). (b) Higher magnification highlighting glassy membrane (arrows). (c) Higher magnification of dermoepidermal junction. Cell-poor interface change with superficial dermal oedema and loss of detail in dermal collagen fibres. Vacuolar degeneration of basal keratinocytes with apoptosis (arrows) and focal subepidermal cleft (arrowhead). Apoptotic keratinocytes are located in basal layer (arrows). Haematoxylin and eosin, ×40 (a) and ×400 (b,c).
All cases meeting these histological inclusion criteria were included in the analysis to define population characteristics. Only cases submitted by veterinarians who were members of the American College of Veterinary Dermatology were analysed for clinical descriptions, treatment and outcome data. This was done because it was assumed that it would provide better consistency in lesion descriptions. For data capture, information was collected from skin biopsy histopathological reports, a written survey and a request for medical records was sent to the submitting clinician for each case (Figure S1).
Case signalment (age, sex, breed), weight, historical or concurrent diseases, current/recent medications, physical examination findings, location and description of lesions, systemic signs (including fever, lethargy, inappetence, weakness, vomiting, diarrhoea, coughing, respiratory effort, lameness, neurological abnormalities, pain, lymphadenopathy and sneezing/nasal discharge), diagnostic tests/laboratory data (complete blood count, biochemical analysis), vaccine history (timing and type of vaccine given), treatments (drugs and doses used) and outcome data (subjective descriptions) were recorded. When known, all medications used for treatment were recorded as dose per day. For the purposes of this study a “vaccination site” was defined as a discrete, unilateral lesion specifically described on the flank/thigh/hip or shoulder/lateral thorax. If a historical or physical examination finding was not noted in the medical record, it was considered to be a missing datum. Cases were excluded from end-point analysis if clinical and historical data were not obtained or if a follow-up period of at least three months was not recorded.
A vaccine scoring rubric was created to evaluate the likelihood of each case being associated with a recent vaccination. (Table 1). Higher numbers indicated higher likelihood of disease being associated with vaccination.
| Score | |
|---|---|
| 0 | No recent vaccine (>6 months), no lesion |
| 1 | Vaccine history (<6 months) + no lesion OR unknown vaccine history + no lesion |
| 2 | Vaccine lesion present + unknown vaccine history |
| 3 | Vaccine lesion present + known vaccine history |
| 4 | Vaccine lesion present + known vaccine history + identification of focal lymphoplasmacytic panniculitis or vaccine material on biopsy |
| 5 | Vaccine lesion present + known vaccine history + identification of vaccine material on biopsy + relapse associated with vaccination |
- A “vaccine lesion” was defined as a focal area of alopecia, hyper- or hypopigmentation, erythema, crust/scale or ulcer on the left or right flank, left or right lateral thigh, left or right shoulder/lateral thorax specifically noted on dermatological examination. Vaccine material within biopsies was identified as blue-grey amorphous material within macrophages or within the dermis/subcutis.
Case outcomes were evaluated using two separate systems (Tables 2 and 3). Both systems were created in an effort to provide the most objective assessment of case severity and to help identify prognostic factors. First, to reflect difficulty in achieving clinical control in each case, a treatment score was created to evaluate the extent of chosen medical therapy. The scoring system was adapted for retrospective purposes from systems previously reported for atopic dermatitis.9-11 A few drugs were added to this scheme (azathioprine, leflunomide) and all doses were scored as the maximum dose that each dog received (Table 2). Treatment scores were cumulative, meaning a higher treatment score indicated the use of a greater number of medications and more potent immunosuppressive agents. Second, in order to best represent long-term outcomes, an end-point scoring system was created based on recorded death/euthanasia (if related to the ischaemic disease) or last recorded medical therapy if the dog was alive at last contact with the client. Cases that died for reasons unrelated to ischaemic disease were scored based on medications being used at last contact before death (Table 3). A higher number indicated a better outcome.
| Group 1 (10 points each) | Group 2 (20 points each) | Group 3 (30 points each) | Group 4 (40 points each) | Group 5 (50 points each) | Group 6 (60 points each) |
|---|---|---|---|---|---|
| Vitamin E | Pentoxifylline | Topical steroids | Oral steroids* (0.5–1 mg/kg) | Oral steroids* (1–2 mg/kg) | Oral steroids* (>2 mg/kg) |
| Fish oil | Doxycycline | Topical tacrolimus | Ciclosporin (<5.5 mg/kg) | Ciclosporin (5.5–7) mg/kg) | Ciclosporin > 7 mg/kg |
| Niacinamide† | Hydroxychloroquine | Azathioprine | |||
| Leflunomide | |||||
| Mycophenolate | |||||
| Dapsone | |||||
| Sulfasalazine |
- *Steroids were recorded as “prednisone equivalent” doses. All doses were recorded in mg/kg/day. If the dose was not recorded for a specific drug, it was scored as the lowest number for that drug.
- †Niacinamide was most often used in combination with doxycycline.
| Score | Category | Criteria |
|---|---|---|
| 1 | Euthanized | Euthanasia related to disease at any point (either disease progression or due to medication adverse effects) |
| 2 | Poor | High doses of steroids, multiple immunosuppressive agents* tried and/or needed long-term |
| 3 | Fair | Immunosuppressive agents* or low-dose steroids (<0.5 mg/kg/day) or topical steroids continued long-term |
| 4 | Good | Vitamin E, fish oil, or pentoxifylline, doxycycline, niacinamide or topical tacrolimus continued long-term |
| 5 | Excellent | Vitamin E, fish oil only with complete resolution, or no medication with stable disease |
| 6 | Complete resolution | No medications |
- End-point scores were determined by the type of medications each patient continued to use at the last contact (more than three months previously) to control their disease.
- *Immunosuppressive agents included ciclosporin, azathioprine, leflunomide, hydroxychloroquine, sulfasalazine and dapsone.
Statistical analysis
A statistical software program (JMP 14.2, SAS Institute Inc.; Cary, NC, USA) was used to generate descriptive statistics for breed, sex, weight and disease characteristics, and examine relationships between selected variables. A one sample proportion test was conducted for each breed within the 177 cases to determine if the proportion differed from the target proportion calculated for the same breeds included in the histopathology database (81,525 of 125,567 total cases). Relationships between case signalments and scoring metrics were analysed using one-way ANOVA. Bivariate analysis was used to determine an association between case signalment and scoring metrics.
Results
Two hundred and eighty-two cases were initially identified from the histopathology database. Of those, 177 met inclusion criteria and were used for breed and signalment data. This equated to 0.14% (177 of 125,567) of all cases where biopsies were submitted during that time period. Of the 177 cases, records and/or surveys were obtained from 93 cases for historical data and analysis.
Signalment
Of the 177 cases, the median age at time of skin biopsy was five years (range 0.33–14 years). There were 78 females (44.6%; 61 spayed, 17 intact) and 95 males (54.8%; 67 neutered and 28 intact), four were unknown sex. There were 43 breeds represented. Mixed breed dogs were the largest group (45 of 177; 25.4%), followed by Chihuahuas (12 of 177; 6.8%), Jack Russell terriers, (12 of 177; 6.8%), toy and miniature poodles, (12 of 177; 6.8%), Yorkshire terriers (seven of 177; 4%), Dachshunds (7 of 177; 4%) and Maltese terriers (five of 177; 2.8%)). All breeds are listed in Table S1. Using the patient population of the biopsy service as a reference, the following breeds were significantly over-represented: Chihuahuas (P < 0.0001), Chinese crested dogs (P < 0.0001), Maltese terriers (P < 0.0001), toy and miniature poodles (P < 0.0001), rat terriers (P < 0.0001), schipperkes (P < 0.0001), Yorkshire terriers (P = 0.0003), toy fox terriers (P = 0.0007) and Jack Russell terriers (P = 0.001). Body weight was recorded in 77 of 93 cases for which additional records were obtained. The median weight was 7.3 (1.32–50.3) kg, with 49 of 77 (63.6%) being <10 kg.
Clinical presentation
Alopecia was the most common lesion noted in 79 of 93 dogs (85%). Other common lesions included crust (65.5%), scale (57%), erythema (44.1%), erosions/ulcers (38.7%) and hyperpigmentation (36.6%). Pruritus also was common, noted in 32 of 93 dogs (34.4%). Other lesions were noted in fewer numbers (Table S2). The median number of lesion sites was four (range one to 13). Thirteen dogs had a single lesion; of these, nine were vaccination site lesions, two were found on pinna, one on the tail and one on the trunk. Excluding vaccine-site-only lesions, dogs with one local lesion had greater body weights (mean = 15.4 kg) and were older (mean = 7.3 years of age) compared to all cases. Overall, lesions were most commonly found on the pinnae in 61 of 93 cases (65.59%), at vaccination sites in 43 (46.3%), and periocular/face in 39 (41.94%) (see Table S3 and Figure 2). Systemic signs were reported in 29 of 93 dogs (31.2%). The most frequently reported were lethargy in 20 of 93 dogs (21.5%), fever in 12 (13%), inappetance in 12 (13%) and lameness in six (6.5%).
Clinical images from two dogs with ischaemic dermatopathy.
(a,b) Case 166: Coalescing foci of scarring alopecia with hyperpigmentation and hypopigmentation in the periocular region and temporal region. Scalloped and notched ear margin with thick adherent scale/crust. (b,c) Case 137: Symmetrical ulcers on the concave pinnae. Annular focus of scarring alopecia and hyperpigmentation on the dorsal calvarium. Hyperpigmentation, erythema and alopecia on the phalanges.
Historical data
Of the 93 cases, 35 (38%) had reported concurrent disease(s) at diagnosis. Recently diagnosed or concurrent medical conditions are shown in Table 4. In 80 cases where concurrent medication use was reported, 25 were being administered medication at the reported time of onset. Medications reported were either antiparasitic preventatives or being used for management of concomitant diseases. Medications included levothyroxine (one case), azathioprine (one), chlorambucil (one), prednisone/prednisolone (five), milk thistle (one), insulin (one), phenobarbital (one), atenolol (one) carprofen (two), meloxicam (one), furosemide (one), hydroxyzine (one), pentoxifylline (one), topical ear medications (four), various antiparasitics [selamectin (one), ivermectin (three), afoxolaner (one), imidacloprid/moxidectin (one) sulfadimethoxine (one)], pain medications [tramadol (one), methocarbamol (one)] and various antibiotics [amoxicillin (one), amoxicillin-clavulanic acid (four), cefovecin (one), clindamycin (one), enrofloxacin (two), metronidazole (one)]. No clinical suspicion was noted for any medications used. Due to incomplete data, medication scores could not be calculated.
| Neurological disease | Intervertebral disc disease (2), seizures (2) |
| Skin and ear disease | Atopy (5), otitis externa (5), demodicosis (2) |
| Infection | Scabies (2), kennel cough (1), anal gland abscess (1), giardia (1) |
| Endocrine disease | Hypothyroidism (2), hyperadrenocorticism (1), diabetes mellitus (1) |
| Neoplasia | Testicular (undefined), t-zone lymphoma |
| Liver disease | Vacuolar hepatopathy/portal fibrosis (1) |
| Auto-immune disease | Immune-mediated haemolytic anaemia/immune-mediated thrombocytopenia (1), sebaceous adenitis (1) |
| Heart disease | Heart murmur (unspecified) (3), heartworm disease (1), pulmonic stenosis (1) |
| Other | Lung lobectomy (1), periodontal disease (3), bladder stones (1), hypoglycaemia (1), hip dysplasia (1) |
- Disease(s) present before onset or within one month of diagnosis of ischaemic dermatopathy. Cases may have had more than one disease listed.
Vaccine scores
The average vaccine score was two. For the purposes of analysis, cases with a vaccine score < 2 (1 or 0) were considered unlikely to have an association with a vaccine. A vaccine score < 2 was calculated for 48 of 93 (51.6%) cases. Higher vaccine scores were positively correlated with higher body weights (P = 0.0029, r2 = 0.96) and younger ages (P = 0.0110, r2 = 0.07). Two lesion sites – the dorsal calvarium (33 observations) and digits (16 observations) – were significantly less likely to be associated with vaccination (mean 1.39, P = 0.003 and mean 1.18, P = 0.0151, respectively). A lower vaccine score was associated with a greater number of systemic signs, but this relationship was not statistically significant (P = 0.116). The effects of individual systemic signs were not evaluated. Vaccine scores were not significantly different between cases with concurrent disease and those without; nor were vaccine scores significantly associated with the total number of lesional sites. Only toy poodles (n = 6) had a significantly higher vaccine score when compared to the case population (mean vaccine score = 4, P = 0.0005).
Treatments
Of the 80 cases for which treatment use was reported, 73 (91.3%) were treated with pentoxifylline. The mean dose used was 47.12 mg/kg/day (range: 18–112.5 mg/kg/day). The next most common drugs used were corticosteroids (in 41 of 80 cases). Of these, 18 of 41 (44%) used a prednisone-equivalent dose greater than 1 mg/kg/day (initial dose). Other common medications included vitamin E in 33 of 93 cases (35.5%) and ciclosporin (mean initial dose: 5.52 mg/kg/day), in 16 of 93 cases (17.2%). See Table S4 for all treatments reported.
Treatment scores
There were 62 cases with follow-up greater than three months which were included in the calculation of end-point scores; however, treatment scores could not be calculated in three of these cases. Of the 59 calculated, the mean treatment score was 82.73, and the median was 60 (range 0–380). Scores of <100 were calculated in 40 of 59 cases (67.8%), whereas 16 cases (27.1%) scored between 100 and 200 and three cases (4.8%) scored >200 (240, 300, 380). When treatment scores were compared to clinical characteristics, only pinnal lesions (n = 42) were significantly associated with higher treatment scores (95.04, P = 0.005). The number of systemic signs was positively correlated with higher treatment scores (P = 0.029, r2 = 0.08). Of all breeds, only miniature pinschers (n = 2) had a significantly higher mean treatment score than other breeds (240, P = 0.0008). Treatment scores did not vary significantly with weight or age at time of biopsy.
End-point scores
End-point scores (1–6) were designated in 62 cases with greater than three months of follow-up or that were euthanized/died at any point. Higher end-point scores were indicative of better outcomes. The average end-point score was 3.35. Half of the cases (31 of 62) achieved an end-point that was considered “good” or better, whereas 21 of these cases (33.9%) were considered “fair” or worse, and remained on at least one immunosuppressive agent for control. Ten of 62 cases (16.1%) were euthanized for issues related to ischaemic disease and five of these were euthanized for severe systemic disease contemporaneous with skin lesion onset. Systemic diseases included aspiration pneumonia, lung lobe torsion and polyarthropathy, unspecified hepatopathy, protein losing enteropathy and unclassified central neurological disease, and protein losing nephropathy (euthanized one year later). Three of 10 cases were euthanized due to lack of response to therapy (2.5–5 months post diagnosis). Two cases died or were euthanized for unknown reasons shortly after diagnosis: one which was euthanized three weeks after starting azathioprine and one which died at home one month after diagnosis.
When end-point scores were compared to clinical presentations, cases with a body weight <10 kg had significantly worse end-point score (mean score = 2.88) than those weighing >10 kg (mean score = 4.14, P = 0.0045). Increasing weight (P = 0.0128, r2 = 0.1) and higher vaccine scores (P = 0.037, r2 = 0.07) were positively correlated with end-point score. Increasing age (r2 = 0.13, P = 0.005), number of lesion sites (r2 = 0.14, P = 0.003) and number of systemic signs (r2 = 0.14, P = 0.0034) were negatively correlated with end-point (see Figure 3). Cases with pinnal lesions had significantly worse end-point scores (mean score = 2.97, P = 0.0033) and higher treatment scores (mean score = 95.04, P = 0.0049). Cases with paw pad lesions had significantly worse end-point scores (mean score = 2.35, P = 0.0005) but not significantly higher treatment scores (P = 0.11). Having only a single vaccine-site lesion trended toward a better end-point score, but was not statistically significant (mean score = 4.6, P = 0.073). All lesions associated with end-points significantly lower than the mean end-point for all cases are reported in Table S3)
Relationship between selected variables and end-point scores for dogs with ischaemic dermatopathy.
End-point scores compared with variables for 62 cases with follow-up greater than three months. Bars within diamond plot for weight distribution indicate the mean ± standard deviation. Continuous variables were compared with a line-of-best-fit. Low body weight (<10 kg), increasing age, number of lesions and number of systemic signs were negatively correlated with end-point (P = 0.0128, P = 0.0048, P = 0.003 and P = 0.0034, respectively) (ANOVA).
The use of pentoxifylline (mean end-point score = 3.27) was not significantly associated with greater end-point score; however, only seven cases were not treated with pentoxifylline. Additionally, there was no significant difference in end-point scores between cases treated with higher (>40 mg/kg/day) or lower (<40 mg/kg/day) doses of pentoxifylline. The use of steroids was associated with a significantly worse end-point score (mean score = 2.72, P = 0.001).
Breed-specific findings
Yorkshire terriers (n = 2) had significantly worse end-point scores (mean score = 1, P = 0.036) but not significantly different treatment scores (mean treatment score = 80) when compared to all other cases. Miniature pinschers (n = 2) had significantly higher treatment scores (mean treatment score = 240, P = 0.0008) but end-point scores were not significantly worse (mean score = 2.5, P = 0.453).
Discussion
Although a few case series have been published examining vaccination-associated ischaemic dermatopathy and a few cases of idiopathic ischaemic disease have been reported, the data presented here represent the first comprehensive description of canine ischaemic dermatopathy.2-5, 12, 13 This study identified a number of breed predispositions and various factors that may be considered prognostic when dealing with cases of ischaemic dermatopathy (exclusive of FDM). Importantly, over half of the cases presented here were considered unlikely to be vaccine-associated, demonstrating the need to investigate other underlying causes for this condition.
The original case series of vaccine-associated ischaemic dermatopathy published in 1986 described 13 dogs (10 of 13 were poodles, two were bichon frises and one was a shih tzu) each with a single lesion at the site of recent rabies vaccination.3 A subsequent case series published in 1999 included three cases of generalized ischaemic lesions following rabies vaccination in a Shetland sheepdog, shih tzu and Pomeranian.2 Four probable cases of vaccine-induced ischaemic disease, with consistent clinical lesions, were included in retrospective analysis of vasculitis in a further study; breeds included two bichon frises, one Maltese terrier and a miniature poodle.12 Altogether, published data, along with clinical experience, suggested that small breed dogs were at increased risk.
In the present study, which included both vaccine-associated and other subtypes of ischaemic disease, small breed dogs represented the majority of cases, with 63.6% being <10 kg. Consistent with previous studies, Chihuahuas, toy and miniature poodles, and Maltese terriers were statistically over-represented compared to the general population. However, a number of other small breeds including Yorkshire terriers, Chinese crested dogs, rat terriers, schipperkes and toy fox terriers also were over-represented in this patient population. Labrador retrievers (n = 6) and boxers (n = 1) were significantly under-represented based on the background population. Smaller body weights were significantly correlated with lower vaccine scores (P = 0.0029, r2 = 0.96), a finding that appears to be at odds with previously published reports where vaccine-associated disease has been described almost exclusively in small breeds.2, 3, 12
Large differences between breeds were identified. As an example, toy poodles had significantly higher vaccine scores (five of six had a vaccine score of ≥4) when compared to Yorkshire terriers (where all four had a vaccine score of ≤1). Additionally, Jack Russell terriers (n = 5) had a low mean vaccine score of 1.33. A previous report described five Jack Russell terriers with identical clinical and histological characteristics to those described in this study. In that series, two of five dogs were very young (three and seven months of age) at initial presentation.13 In the Jack Russell terriers reported here, the mean age at skin biopsy (4.16 years) was not significantly different from the group as a whole; however, several of these cases had lesions described since puppyhood. This may represent a significant breed-specific predisposition to juvenile-onset ischaemic disease/vasculitis. Taking this into account, the association between lower vaccines scores and smaller body weights may be explained by greater heterogeneity in the types of ischaemic disease affecting small-breed dogs as compared to large-breed dogs.
Based on the findings of this study, along with data published previously, there are clearly breeds at higher risk and lower risk of developing ischaemic disease.2, 3, 12 Two candidate genes (PAN2 and MAP3K7CL) have been identified using whole-genome sequencing in collies and Shetland sheepdogs with FDM. Polymorphisms in these two genes, along with corresponding risk MHC haplotypes, conferred increased susceptibility to disease.14 These findings also shed some light on the differences in the age of onset in dogs with FDM; collies with a moderate-risk genotype were more likely to develop the disease later in life than individuals with a high-risk genotype. Notably, these polymorphisms were documented in other breeds, some of which were over-represented in the present study. Breeds reported to have of one or more polymorphisms, either in the two new candidate genes or MHC haplotypes associated with FDM, included the following: Chihuahuas, fox terriers, Jack Russell terriers, poodles, Labrador retrievers, Yorkshire terriers, Dachshunds, Cardigan Welsh corgis and cairn terriers. However, moderate-risk genotypes were identified in only two breeds, Jack Russell terriers and Cardigan Welsh corgis.14 These findings seem to strongly suggest a genetic susceptibility in the development of ischaemic disease. Further studies are needed to correlate these genetic findings with disease in at-risk breeds to ascertain pathogenesis and heritability, in North America and other regions.
Currently, little is known about the pathogenesis of ischaemic dermatopathy. The most well-studied of ischaemic diseases, FDM, is thought to be driven by an immune-mediated mechanism in genetically susceptible individuals following an environmental trigger. Environmental triggers such as infections, vaccines, drugs, UV exposure and stressful events have been described anecdotally in connection with individual cases.15 In the original series of focal vaccine-associated ischaemic disease, a direct role of the vaccine substance in the development of disease was suspected. Dogs in that study all had lesions located only at the site of rabies vaccination.3 Interestingly, mirroring the population in that publication, the current study found that of nine dogs with a vaccine-site-only lesion, three were toy poodles, one was a bichon frise and one a shih tzu, indicating a possible propensity for these breeds to develop local, but not generalized disease. Additionally, although not statistically significant, cases in this study with focal lesions only, had lower treatment scores and higher end-point scores than the group as a whole, indicating that localized disease may have a better overall prognosis. A subsequent study examining dogs with vaccine-associated multifocal disease described hypovascularity and complement deposition around microvasculature, suggesting a role for the innate immune response in development of disease.2 Although the pathogenesis remains enigmatic, human research on juvenile dermatomyositis suggests that complement does play a role in damage to small blood vessels in the skin and muscles of affected patients.16-18
Although vaccines appear to play a role in some cases, generalized idiopathic ischaemic dermatopathy more likely encompasses a diverse group of diseases with variations in severity and unidentified but wide-ranging triggers. Because the majority of cases (48 of 93) represented in this dataset were likely to be unrelated to vaccination, it is clear that vaccines are only part of the puzzle. This study did not find any significant association with underlying disease, nor did it identify other triggers. Consequently, additional studies will be needed to examine specific subsets of the disease with the goal of identifying triggers and understanding the pathogenesis of ischaemic lesions.
The outcome data reported here reveal a number of trends that may be used as prognostic factors and help to guide therapy for patients. In general, half of the cases had high end-point scores, indicating that a large number of cases had a good or better outcome and could be maintained on relatively safe medications long-term. However, the other half of the cases either died, were euthanized, or continued long-term immunosuppressive therapies. Factors associated with lower (worse) end-point scores included weights <10 kg, increasing age, greater number of lesion sites, presence of systemic signs and specific lesion sites including the pinna and paw pad. It is possible that lesions at these sites were perceived as more severe by owners due to pinnal bleeding and/or lameness. This perception may have resulted in increased propensity for owners to continue long-term treatment in dogs with these lesions, whereas alopecic or atrophic lesions at other sites may have been ignored. Older dogs may have had worse outcomes due to the presence of comorbidities or may not have been treated as aggressively due a perception of increased risk for adverse effects and fear that they may have fared worse than younger dogs. Interestingly, because lower vaccine scores were correlated with older ages, this is likely to indicate a difference in pathogenesis for this subgroup.
Treatment scores were relatively low overall (median treatment score = 60) with just a few outliers. This finding indicates that remission or stable disease may have been achieved without significant changes in medications or combination therapy. Based on treatment of dermatomyositis, pentoxifylline historically has been used as the mainstay of therapy for its presumed rheological and immunomodulatory effects; however, few reports have evaluated its mechanism of action in dogs, especially as it pertains to use for ischaemic disease. One study examining pharmacokinetics in dogs did not find any measurable haematological changes, indicating that the immunological effects may play a larger role in its effect.19 This study did not identify differences in outcome between dogs treated with pentoxifylline and those that were not. However, interpretation of this finding is difficult because the majority of cases (86 of 93) and all cases with an end-point calculation (62 of 62) were treated with this medication. Additionally, there was a large variation in the dose of pentoxifylline that was employed, making generalizations difficult. Thus, further evaluation of the mechanism of action of this drug in dogs and its use in prospective, controlled studies, are needed to more closely investigate the effect of pentoxifylline on disease outcome.
Due to its retrospective nature, this study has a number of inherent limitations which include variations in medical record reporting, completeness and lack of follow-up. Foremost, because FDM was excluded based on reported histories and breed, it is possible that some cases that could be more consistent with FDM, but in atypical or unreported breeds, were included. Furthermore, although records were reviewed in most included cases, timelines associated with disease onset (e.g. time between vaccination and onset of lesions) and progression were difficult to interpret due to medical record quality and the possibility of subtle lesions being missed at disease onset. Clinical outcomes also were difficult to establish based on records because lesions may be slow to progress and others result in scarring that may never fully resolve. Additionally, many records were not complete in their description of progression or resolution of lesions. For these reasons, the end-point score was created to provide more objective end-point evaluation but it may not reflect a “final” outcome. It should be noted that the treatment scoring system employed does not reflect a static time period. Thus, dogs with a longer follow-up have the potential to have a greater treatment score. However, this score depicts difficulty in achieving remission in cases that may not have been reflected in the final outcome. Although this report, to the best of the authors’ knowledge, contains the largest number of dogs with ischaemic disease to date, there are still relatively few individuals representing various breeds. This complicates interpretation of breed-specific susceptibilities. Lastly, this study, along with the majority of previous publications, was based on canine populations in the USA/North America, and thus findings may not be transferrable to other areas of the world.2, 3, 12-14
Future studies are needed to help identify triggers of idiopathic disease and further elucidate disease pathogenesis, including the role of specific vaccine components. Given the newly identified genotypes associated with disease severity in cases of FDM, examination of genetic factors may help to understand why certain breeds are at increased risk.
Acknowledgements
The authors would like to thank Kathyrn A. Rook and Michael H. Goldschmidt for guidance and support.
